In [1]:
import pandas as pd
import seaborn as sns
from sklearn.preprocessing import PolynomialFeatures

特徴量エンジニアリング¶

日付¶

In [2]:
# データ準備
df = pd.read_csv('bike_share.csv')
df
Out[2]:
datetime season holiday workingday weather temp atemp humidity windspeed casual registered count
0 2011-01-01 00:00:00 1 0 0 1 9.84 14.395 81 0.0000 3 13 16
1 2011-01-01 01:00:00 1 0 0 1 9.02 13.635 80 0.0000 8 32 40
2 2011-01-01 02:00:00 1 0 0 1 9.02 13.635 80 0.0000 5 27 32
3 2011-01-01 03:00:00 1 0 0 1 9.84 14.395 75 0.0000 3 10 13
4 2011-01-01 04:00:00 1 0 0 1 9.84 14.395 75 0.0000 0 1 1
... ... ... ... ... ... ... ... ... ... ... ... ...
10881 2012-12-19 19:00:00 4 0 1 1 15.58 19.695 50 26.0027 7 329 336
10882 2012-12-19 20:00:00 4 0 1 1 14.76 17.425 57 15.0013 10 231 241
10883 2012-12-19 21:00:00 4 0 1 1 13.94 15.910 61 15.0013 4 164 168
10884 2012-12-19 22:00:00 4 0 1 1 13.94 17.425 61 6.0032 12 117 129
10885 2012-12-19 23:00:00 4 0 1 1 13.12 16.665 66 8.9981 4 84 88

10886 rows × 12 columns

カラムの説明
datetime - hourly date + timestamp
season - 1 = spring, 2 = summer, 3 = fall, 4 = winter holiday - whether the day is considered a holiday
workingday - whether the day is neither a weekend nor holiday
weather -
1: Clear, Few clouds, Partly cloudy, Partly cloudy
2: Mist + Cloudy, Mist + Broken clouds, Mist + Few clouds, Mist
3: Light Snow, Light Rain + Thunderstorm + Scattered clouds, Light Rain + Scattered clouds
4: Heavy Rain + Ice Pallets + Thunderstorm + Mist, Snow + Fog
temp - temperature in Celsius
atemp - "feels like" temperature in Celsius
humidity - relative humidity
windspeed - wind speed
casual - number of non-registered user rentals initiated
registered - number of registered user rentals initiated
count - number of total rentals

In [3]:
# 日付カラムの確認
df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 10886 entries, 0 to 10885
Data columns (total 12 columns):
 #   Column      Non-Null Count  Dtype  
---  ------      --------------  -----  
 0   datetime    10886 non-null  object 
 1   season      10886 non-null  int64  
 2   holiday     10886 non-null  int64  
 3   workingday  10886 non-null  int64  
 4   weather     10886 non-null  int64  
 5   temp        10886 non-null  float64
 6   atemp       10886 non-null  float64
 7   humidity    10886 non-null  int64  
 8   windspeed   10886 non-null  float64
 9   casual      10886 non-null  int64  
 10  registered  10886 non-null  int64  
 11  count       10886 non-null  int64  
dtypes: float64(3), int64(8), object(1)
memory usage: 1020.7+ KB
In [4]:
# datetimeカラムを日付のデータタイプに変換
df['datetime'] = pd.to_datetime(df['datetime'])
In [5]:
# datetimeカラムのDtypeがdatetimeになっていることを確認
df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 10886 entries, 0 to 10885
Data columns (total 12 columns):
 #   Column      Non-Null Count  Dtype         
---  ------      --------------  -----         
 0   datetime    10886 non-null  datetime64[ns]
 1   season      10886 non-null  int64         
 2   holiday     10886 non-null  int64         
 3   workingday  10886 non-null  int64         
 4   weather     10886 non-null  int64         
 5   temp        10886 non-null  float64       
 6   atemp       10886 non-null  float64       
 7   humidity    10886 non-null  int64         
 8   windspeed   10886 non-null  float64       
 9   casual      10886 non-null  int64         
 10  registered  10886 non-null  int64         
 11  count       10886 non-null  int64         
dtypes: datetime64[ns](1), float64(3), int64(8)
memory usage: 1020.7 KB
In [6]:
# 日付データからそれぞれの値を取得
df['year'] = df['datetime'].dt.year
df['month'] = df['datetime'].dt.month
df['dayofyear'] = df['datetime'].dt.dayofyear
df['dayofweek'] = df['datetime'].dt.dayofweek
df['weekofyear'] = df['datetime'].dt.isocalendar().week
df['quarter'] = df['datetime'].dt.quarter
df['isleap'] = df['datetime'].dt.is_leap_year

多項式特徴量¶

In [7]:
# データロード
df = pd.read_csv('penguins_size.csv')
df.dropna(inplace=True)
# 多項式特徴量作成
poly = PolynomialFeatures(degree=2, include_bias=False)
df[['culmen_length_mm', 'culmen_depth_mm','culmen_length_mm^2', 'culmen_length_mm * culmen_depth_mm', 'culmen_depth_mm^2' ]] = poly.fit_transform(df[['culmen_length_mm', 'culmen_depth_mm']])
sns.pairplot(df, hue='species')
Out[7]:
<seaborn.axisgrid.PairGrid at 0x405fdc94f0>

binning¶

In [8]:
df['body_mass_g_bin'] = pd.cut(df['body_mass_g'], bins=10, labels=False)
# カテゴリカル変数として扱う場合dtypeをobjectに変換．数値変数として扱ってもよい
df['body_mass_g_bin'] = df['body_mass_g_bin'].astype('object')
df.info()

<class 'pandas.core.frame.DataFrame'>
Int64Index: 334 entries, 0 to 343
Data columns (total 11 columns):
 #   Column                              Non-Null Count  Dtype  
---  ------                              --------------  -----  
 0   species                             334 non-null    object 
 1   island                              334 non-null    object 
 2   culmen_length_mm                    334 non-null    float64
 3   culmen_depth_mm                     334 non-null    float64
 4   flipper_length_mm                   334 non-null    float64
 5   body_mass_g                         334 non-null    float64
 6   sex                                 334 non-null    object 
 7   culmen_length_mm^2                  334 non-null    float64
 8   culmen_length_mm * culmen_depth_mm  334 non-null    float64
 9   culmen_depth_mm^2                   334 non-null    float64
 10  body_mass_g_bin                     334 non-null    object 
dtypes: float64(7), object(4)
memory usage: 31.3+ KB

四則演算¶

In [9]:
df['culmen_diff'] = df['culmen_length_mm'] - df['culmen_depth_mm']
df['culmen_ratio'] = df['culmen_length_mm'] / df['culmen_depth_mm']
%matplotlib inline
sns.pairplot(df, hue='species')
Out[9]:
<seaborn.axisgrid.PairGrid at 0x40625cb250>

集約値¶

In [10]:
# データロード
df_app = pd.read_csv('homecredit_application.csv')
df_prev_app = pd.read_csv('homecredit_previous_application.csv')
In [11]:
# 今回のcredit申請データ
df_app.head(3)
Out[11]:
SK_ID_CURR TARGET NAME_CONTRACT_TYPE CODE_GENDER FLAG_OWN_CAR FLAG_OWN_REALTY CNT_CHILDREN AMT_INCOME_TOTAL AMT_CREDIT AMT_ANNUITY ... FLAG_DOCUMENT_18 FLAG_DOCUMENT_19 FLAG_DOCUMENT_20 FLAG_DOCUMENT_21 AMT_REQ_CREDIT_BUREAU_HOUR AMT_REQ_CREDIT_BUREAU_DAY AMT_REQ_CREDIT_BUREAU_WEEK AMT_REQ_CREDIT_BUREAU_MON AMT_REQ_CREDIT_BUREAU_QRT AMT_REQ_CREDIT_BUREAU_YEAR
0 100002 1 Cash loans M N Y 0 202500.0 406597.5 24700.5 ... 0 0 0 0 0.0 0.0 0.0 0.0 0.0 1.0
1 100003 0 Cash loans F N N 0 270000.0 1293502.5 35698.5 ... 0 0 0 0 0.0 0.0 0.0 0.0 0.0 0.0
2 100004 0 Revolving loans M Y Y 0 67500.0 135000.0 6750.0 ... 0 0 0 0 0.0 0.0 0.0 0.0 0.0 0.0

3 rows × 122 columns

In [12]:
# 以前のcredit申請データ
df_prev_app.head(3)
Out[12]:
SK_ID_PREV SK_ID_CURR NAME_CONTRACT_TYPE AMT_ANNUITY AMT_APPLICATION AMT_CREDIT AMT_DOWN_PAYMENT AMT_GOODS_PRICE WEEKDAY_APPR_PROCESS_START HOUR_APPR_PROCESS_START ... NAME_SELLER_INDUSTRY CNT_PAYMENT NAME_YIELD_GROUP PRODUCT_COMBINATION DAYS_FIRST_DRAWING DAYS_FIRST_DUE DAYS_LAST_DUE_1ST_VERSION DAYS_LAST_DUE DAYS_TERMINATION NFLAG_INSURED_ON_APPROVAL
0 1891648 100373 Consumer loans 10872.990 115695.0 115119.0 11569.5 115695.0 SATURDAY 22 ... Industry 12.0 low_normal POS household with interest 365243.0 -1013.0 -683.0 -713.0 -709.0 0.0
1 2670402 100077 Cash loans NaN 0.0 0.0 NaN NaN WEDNESDAY 14 ... XNA NaN XNA Cash NaN NaN NaN NaN NaN NaN
2 1514511 101011 Consumer loans 6856.065 65866.5 66735.0 6588.0 65866.5 FRIDAY 21 ... Consumer electronics 12.0 middle POS household with interest 365243.0 -596.0 -266.0 -296.0 -289.0 0.0

3 rows × 37 columns

In [13]:
df_prev_app.sort_values('SK_ID_CURR').head()
Out[13]:
SK_ID_PREV SK_ID_CURR NAME_CONTRACT_TYPE AMT_ANNUITY AMT_APPLICATION AMT_CREDIT AMT_DOWN_PAYMENT AMT_GOODS_PRICE WEEKDAY_APPR_PROCESS_START HOUR_APPR_PROCESS_START ... NAME_SELLER_INDUSTRY CNT_PAYMENT NAME_YIELD_GROUP PRODUCT_COMBINATION DAYS_FIRST_DRAWING DAYS_FIRST_DUE DAYS_LAST_DUE_1ST_VERSION DAYS_LAST_DUE DAYS_TERMINATION NFLAG_INSURED_ON_APPROVAL
2547 1038818 100002 Consumer loans 9251.775 179055.0 179055.0 0.0 179055.0 SATURDAY 9 ... Auto technology 24.0 low_normal POS other with interest 365243.0 -565.0 125.0 -25.0 -17.0 0.0
3507 2396755 100003 Consumer loans 6737.310 68809.5 68053.5 6885.0 68809.5 SATURDAY 15 ... Consumer electronics 12.0 middle POS household with interest 365243.0 -2310.0 -1980.0 -1980.0 -1976.0 1.0
1654 1810518 100003 Cash loans 98356.995 900000.0 1035882.0 NaN 900000.0 FRIDAY 12 ... XNA 12.0 low_normal Cash X-Sell: low 365243.0 -716.0 -386.0 -536.0 -527.0 1.0
2934 2636178 100003 Consumer loans 64567.665 337500.0 348637.5 0.0 337500.0 SUNDAY 17 ... Furniture 6.0 middle POS industry with interest 365243.0 -797.0 -647.0 -647.0 -639.0 0.0
2669 1564014 100004 Consumer loans 5357.250 24282.0 20106.0 4860.0 24282.0 FRIDAY 5 ... Connectivity 4.0 middle POS mobile without interest 365243.0 -784.0 -694.0 -724.0 -714.0 0.0

5 rows × 37 columns

In [14]:
# 集約値計算
credit_prev_sum = df_prev_app.groupby('SK_ID_CURR').sum()[['AMT_CREDIT']]
In [15]:
# SK_ID_CURRをキーに表を結合
df_app.join(credit_prev_sum, on='SK_ID_CURR', rsuffix='_PREV_SUM')
Out[15]:
SK_ID_CURR TARGET NAME_CONTRACT_TYPE CODE_GENDER FLAG_OWN_CAR FLAG_OWN_REALTY CNT_CHILDREN AMT_INCOME_TOTAL AMT_CREDIT AMT_ANNUITY ... FLAG_DOCUMENT_19 FLAG_DOCUMENT_20 FLAG_DOCUMENT_21 AMT_REQ_CREDIT_BUREAU_HOUR AMT_REQ_CREDIT_BUREAU_DAY AMT_REQ_CREDIT_BUREAU_WEEK AMT_REQ_CREDIT_BUREAU_MON AMT_REQ_CREDIT_BUREAU_QRT AMT_REQ_CREDIT_BUREAU_YEAR AMT_CREDIT_PREV_SUM
0 100002 1 Cash loans M N Y 0 202500.0 406597.5 24700.5 ... 0 0 0 0.0 0.0 0.0 0.0 0.0 1.0 179055.0
1 100003 0 Cash loans F N N 0 270000.0 1293502.5 35698.5 ... 0 0 0 0.0 0.0 0.0 0.0 0.0 0.0 1452573.0
2 100004 0 Revolving loans M Y Y 0 67500.0 135000.0 6750.0 ... 0 0 0 0.0 0.0 0.0 0.0 0.0 0.0 20106.0
3 100006 0 Cash loans F N Y 0 135000.0 312682.5 29686.5 ... 0 0 0 NaN NaN NaN NaN NaN NaN 2625259.5
4 100007 0 Cash loans M N Y 0 121500.0 513000.0 21865.5 ... 0 0 0 0.0 0.0 0.0 0.0 0.0 0.0 999832.5
... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
995 101152 0 Cash loans F N N 0 112500.0 495985.5 17946.0 ... 0 0 0 0.0 0.0 0.0 0.0 0.0 5.0 2476305.0
996 101153 0 Cash loans F N Y 0 225000.0 1113840.0 57001.5 ... 0 0 0 0.0 0.0 0.0 0.0 0.0 0.0 41850.0
997 101154 0 Cash loans F Y Y 0 144000.0 517536.0 28206.0 ... 0 0 0 0.0 0.0 0.0 0.0 0.0 3.0 478867.5
998 101155 0 Cash loans M N Y 0 315000.0 1288350.0 37800.0 ... 0 0 0 0.0 0.0 0.0 1.0 1.0 1.0 269901.0
999 101156 0 Cash loans M Y Y 2 180000.0 679500.0 27076.5 ... 0 0 0 0.0 0.0 0.0 0.0 0.0 3.0 2115810.0

1000 rows × 123 columns